Cognitive Issues Related to Advanced Cockpit Displays: Supporting the Transition Between

MIT

ICAT

MIT

ICAT

C

OGNITIVE

I

SSUES

R

ELATED TO

A

DVANCED

C

OCKPIT

D

ISPLAYS

:

S

UPPORTING THE

T

RANSITION BETWEEN

I

NTERNAL AND

E

XTERNAL

G

UIDANCE

Maura C. Lohrenz

Report No. ICAT-2003-3

June 2003

MIT International Center for Air Transportation

Department of Aeronautics & Astronautics

Massachusetts Institute of Technology

Cambridge, MA 02139 USA

C

I

R

A

C

D

:

S

T

I

E

G

BY

MAURA C. LOHRENZ

A

A critical issue in military aviation is the pilot’s ability to transition between primarily internal (head-down, instrument-driven) and external (head-up, out of the cockpit) guidance.

Experimental cockpit displays were designed and tested for how well they might support this transition phase for military pilots performing time-critical air-to-ground targeting missions such as Forward Air Control and Close Air Support. Twelve subjects performed three sets of

experiments using a flight simulator (with simulated heads-up display in the forward field of view) connected to a moving-map display. The experiments were designed to help explain which visual cues in the displays might best help a pilot 1) navigate to a given target area (the “flight guidance” phase of a mission) and 2) search for, find and identify a target (the “target acquisition” phase).

In one experiment, subjects flew a mission using three different versions of a plan-view moving-map display: 1) a detailed, topographic moving-map with high-contrast mission overlays, 2) the moving-map only, and 3) the overlays only. Overall, both flight guidance and target acquisition performance were best and workload was lowest while using the simplest, overlays-only display; performance and workload were equivalent or worse with the combination display, and significantly worse with the map-only display. These results suggest that the simplest possible display, portraying only the necessary information, is optimal. The distraction of display clutter often outweighs the potential benefits of additional information, especially when the display must support a time-critical task such as air-to-ground targeting.

In all experiments, subjects demonstrated an increased reliance on internal cues (such as the moving-map and other flight instrumentation) during flight guidance and shifted their attention to the external view outside their virtual cockpit during target acquisition. If the target had moved from its expected location, subjects continued switching focus between internal and external cues (a process known as navigational checking) to correlate the various sources of information and find the target.

This document is based on the thesis of Maura C. Lohrenz submitted to the Department of Aeronautics and Astronautics at the Massachusetts Institute of Technology in partial fulfillment of the requirements for the degree of Master of Science in Aeronautics and Astronautics.

A

This work was sponsored by the Naval Research Laboratory Select Graduate Training Program. I thank my supervisors at NRL, Dr. Herbert Eppert and Mr. Mike Harris, for allowing me the tremendous opportunity to attend MIT and complete my Masters’ degree in Aeronautics and Astronautics.

I am grateful to my MIT advisor, Professor John Hansman, for his guidance, for keeping me focused on this project, and for encouraging me to “think outside the box!”

I wish to thank the F/A-18 pilots at the Naval Air Weapons Station in Patuxent River, MD, for taking the time to talk with me about their missions and digital mapping requirements. I thank my colleagues at the Naval Research Laboratory, Stennis Space Center, who enthusiastically agreed to participate in my experiments. Thank you also to Lt. Joseph Cohn, chair of the NRL Human Subjects Institutional Review Board (HS-IRB), for his assistance in ensuring that the experiments were HS-IRB compliant and for expediting my request for HS-IRB approval. Joshua Pollack (MIT Aeronautics and Astronautics Department) and Stephanie Edwards (NRL Code 7440.1 Moving-map Capabilities Team) contributed considerable time and effort to developing software that links the flight simulator output variables with a real-time moving-map system, without which the experiments in this thesis would not have been possible.

Finally, I thank my husband, Steve, and our children, Erin and Ryan, for their love and support throughout this adventure.

T

C

ABSTRACT...3

ACKNOWLEDGEMENTS...5

TABLE OF CONTENTS...7

TABLES AND FIGURES...9

ABBREVIATIONS...12

1. INTRODUCTION...14

1.1 Motivation...14

1.2 Cockpit moving-maps in support of air-to-ground missions...14

1.3 Overview of thesis organization...17

2. PILOT INTERVIEWS...19

2.1 General Information...19

2.2 Air-to-Ground Missions...21

2.3 Situation Awareness (SA)...25

2.4 Advanced Moving-Map Display Recommendations...27

2.4.1 Geospatial features of interest...27

2.4.2. Correlating ground and air perspectives...29

2.4.3. Timeliness, accuracy and trust...29

2.5 Pilot Flexibility...31

2.6 Supporting the Transition Between Internal and External Guidance...32

3. COGNITIVE PROCESS MODEL...35

4. EXPERIMENTS...38

4.1 Introduction...38

4.2 Methods...40

4.2.1 Participants...40

4.2.2 Apparatus...40

4.2.3 Flight Simulator Settings...41

4.2.4 Targets...44

4.2.5 Moving-Map Settings...45

4.2.6 Procedure...46

4.2.7.1 Performance 48

4.2.7.2 Workload 50

4.3 Experiment A: Route with turn vs. Straight route...51

4.3.1 Overview...51

4.3.2 Hypotheses...51

4.3.3 Results...52

4.3.3.1 Baseline measurements for subjects 52 4.3.3.2 Performance results 54 4.3.3.3 Workload results 55 4.3.3.4 Guidance Cues 55 4.4 Experiment B: Relative benefits of topographic map and mission symbology...57

4.4.1 Overview...57 4.4.2 Hypotheses...58 4.4.3 Results...60 4.4.3.1 Performance results 60 4.4.3.2 Workload results 63 4.4.3.3 Guidance Cues 63 4.5 Experiment C: Acquiring target with HUD vs. without HUD...65

4.5.1 Overview...65 4.5.2 Hypotheses...66 4.5.3 Results...68 4.5.3.1 Performance Results 68 4.5.3.2 Workload Results 72 4.5.3.3 Guidance Cues 74 5. CONCLUSIONS...75 REFERENCES...78

APPENDIX A. Flight simulator to GPS (fs2gps) program source code...83

APPENDIX B. RMSE calculations of flight path, altitude, and airspeed deviations..100

APPENDIX C. Experimental results by subject...104

Tables and Figures

Table 2-1. List of questions used to conduct focused interviews with F/A-18 pilots...20

Table 2-2. Military flight experience of interviewed pilots...21

Table 4-1. Order of runs for Experiment B...57

Table 4-2. Order of runs for Experiment C...66

Figure 1-1. Illustration of a sample FAC / CAS mission...23

Figure 3-1. Primary channels of information used by pilot during air-to-ground missions...35

Figure 3-2. Cognitive Process Model...36

Figure 4-1. Experimental setup...41

Figure 4-2. FS2002 scenery (baseline level of detail) with instrument readout and compass....42

Figure 4-3. Second level of scene detail, with flight path, FAC and target locations...42

Figure 4-4. Third level of detail shows directional indicator...43

Figure 4-5. Current HUD in F/A-18 cockpit...44

Figure 4-6. Three targets (tank, jeep, fuel truck) and false target (car) used in FS2002 scenes.44 Figure 4-7. FalconView moving-map of Kauai overlaid with experimental symbology...45

Figure 4-8. Decision tree used to gauge workload...48

Figure 4-9. Flight guidance performance...49

Figure 4-10. Deviation in altitude (DAp) for each recorded point along the flight path...50

Figure 4-11. Location of routes for Experiment A...51

Figure 4-12. Average performance measurements and workload assessments, by subject...53

Figure 4-13. Performance results from Experiment A...54

Figure 4-14. Workload results from Experiment A...54

Figure 4-15. Percentage of subjects reporting the use of cues in Experiment A...56

Figure 4-16. Three test cases for Experiment B...58

Figure 4-17. Three route / target combinations used in Experiment B...59

Figure 4-18. Summary of performance results from Experiment B...60

Figure 4-19. Summary of workload results from Experiment B...62

Figure 4-20. Percentage of subjects reporting the use of cues in Experiment B...64

Figure 4-22. Flight guidance performance results of Experiment C...68

Figure 4-23. Stationary target acquisition performance results of Experiment C...69

Figure 4-24. Moving target acquisition performance results of Experiment C...70

Figure 4-25. Average number of missed attempts to find target...70

Figure 4-26. Flight guidance workload results of Experiment C...71

Figure 4-27. Target acquisition workload results of Experiment C...73

A

2D / 3D 2 dimensional / 3 dimensional CAS Close Air Support

CLOS Clear Line of Sight

CSAR Combat Search and Rescue DTED Digital Terrain Elevation Data ERF Ego-centered Reference Frame FAC Forward Air Controller

FAC-A FAC-Airborne (i.e., pilot)

FAC-G FAC-Ground (i.e., ground soldier) FEBA Forward Edge of Battle Area FOV Field of View

FFOV Forward Field of View

FLIR Forward-Looking Infrared Radar FLOT Forward Line of Troops

FS2002 Flight Simulator 2002 (Microsoft) GIDB Geospatial Information Data Base GPS Global Positioning System

HAT Height above Terrain (or Threshold or Target) HSI Horizontal Situation Indicator

HUD Head-Up Display IFR Instrument Flight Rules INS Inertial Navigation System

NIMA National Imagery and Mapping Agency NMEA National Marine Electronics Association NPFPS Navy Portable Flight Planning System NRL Naval Research Laboratory

NVG Night Vision Goggles RMSE Root Mean Squared Error SA Situation Awareness

TAMMAC Tactical Aircraft Moving Map Capability TAWS Terrain Avoidance Warning System USGS United States Geological Survey WRF World-centered Reference Frame

1. I

1.1 Motivation

A critical issue in military aviation is the pilot’s ability to transition between primarily internal (head-down, instrument-driven) and external (head-up, out of the cockpit) guidance. Military pilots report difficulties during this transition phase, particularly during time-critical target acquisition missions such as Close Air Support (CAS) and Forward Air Control (FAC). Parallel issues exist in civilian aviation; for example, during circling Instrument Flight Rules (IFR) approaches, the pilot must transition between internal guidance while flying through clouds and external guidance below the ceiling. The military scenario is complicated by the fact that the target may be mobile and unpredictable, whereas the civilian “target” – the runway – is fixed.

The goal of this thesis is to investigate, test, and compare selected cockpit display techniques to improve this transition between internal and external guidance. Tools to synchronize internal and external information paths are expected to support communication and coordination among mission subjects (e.g., CAS pilots and FAC ground troops; or civilian pilots and air traffic controllers), improving the transition phase and increasing the potential for mission success. Three experiments were performed in an attempt to address these issues. Performance and workload measurements were compared for subjects flying a simulated aircraft while referring to conformal, scene-linked, 3-dimensional (3D) symbology on a head-up display (HUD); the equivalent symbology presented on a head-down, 2D moving-map display; or both. Specific questions to be investigated include the following: 1) Which of the tested display techniques best support the transition between internally dominant and externally dominant guidance? 2) Which display techniques best support flight guidance and target acquisition tasks? 3) How do the different display techniques impact performance and workload?

1.2 Cockpit moving-maps in support of air-to-ground missions

Potter, et al. (2002) describe the importance of clearly defining the goals and scope of a display in terms of the cognitive tasks it is intended to support. To that end, designers should specify the decisions to be made, the cognitive performance objectives to be achieved, and the information

(not just the data) to be conveyed by the display. In other words, it is not only important to specify what will be displayed, but how best to display the information.

Navy air-to-ground missions require precise flight guidance to and from the target area (ingress, egress) and timely, accurate target acquisition (including searching for, locating, and positively identifying the target). The flight guidance and target acquisition phases are closely interrelated. Errors during flight guidance may result in a failure to acquire the target; conversely, flight guidance itself can be seen as a series of intermediate target acquisitions as the pilot searches for and finds “lead-in” features along the route, confirming that he or she is on the right track (Conejo and Wickens, 1997).

Flight guidance is supported by a map or electronic map display (with a focus on sources of information inside the cockpit), while target acquisition relies on up-to-date intelligence

information – often communicated by a FAC – and considerable attention and focus on the scene

outside the cockpit. Thus, the pilot must alternate between primarily internal and primarily

external guidance throughout the mission, with the focus on the external “real-world” view commanding more and more attention as the target area approaches. Perhaps more importantly, both phases of the mission (flight guidance and target acquisition) require the pilot to correlate inside (e.g., map) and outside views to ensure mission success.

Many researchers have investigated the ways in which pilots correlate different views or frames of reference. Wickens et al. (1989), Aretz (1991), and others have defined the concept of navigational awareness as a cognitive correlation between an ego-centered reference frame (ERF), established by the pilot’s forward field of view (FFOV), and a world-centered reference frame (WRF), established by a map or electronic map display. They ascertain that a pilot’s inability to correlate these reference frames will result in a lack of navigational awareness, leading to spatial disorientation. Aretz (1991) has identified the cognitive operations required to achieve navigational awareness, including triangulation (in which the ERF must uniquely specify a location in the WRF) and mental rotation (including both circular and forward rotations). This research has shown that circular mental rotation can be aided by presenting the map in a track-up

orientation, such that the pilot’s view of the map corresponds with his or her view of the ground below. ERF-based navigation tasks, such as localization, are best supported by a track-up orientation, while WRF-based tasks, such as pre-mission planning, may be best supported by a north-up orientation. On the other hand, Olmos et al. (1997) has demonstrated that global situation awareness (required for WRF tasks) might be better supported by a track-up map display than a north-up map. Pilots in this study could identify a target's location in the FFOV more quickly with a track-up map than a north-up map, which the authors attributed to a common ERF shared between the FFOV and the rotating map.

Forward mental rotation can be aided by presenting salient geospatial cues (such as target location) in a HUD, such that the pilot’s view of the cues is coincident with the view outside. The high speed of ingress during air-to-ground attack missions leaves little time for a pilot to compare features on a map with features on the ground (Pringle et al., 1998). Rendering the map (or geospatial cues) in a way that is most congruent with the pilot’s forward field of view should assist the pilot during time-constrained tasks such as target acquisition.

Researchers have identified a potentially dangerous consequence of using HUDs known as attentional tunneling, or attending to the information presented on the HUD at the expense of the view outside the cockpit window (e.g., Foyle et al., 1995; Wickens and Long, 1994 and 1995; Hooey, et al., 2001). This problem can be alleviated by replacing superimposed symbology (in which the symbols are mostly presented at a fixed, specific location on the HUD screen) with scene-linked, conformal symbology (in which the symbols are virtually placed in the outside scene and appear to be part of the environment).

Aretz (1998) conducted an extensive literature review of the advantages and disadvantages associated with various frames of reference with respect to electronic map displays for low altitude, visual flight rules (VFR) helicopter flight. Although the current thesis focuses on fixed-wing aircraft, Aretz’s paper is particularly relevant, since many air-to-ground military missions are performed at low altitude and require VFR (or near-VFR) weather conditions for successful target acquisition. Aretz has identified several parameters that should be considered when

determining which reference frame is best for any electronic map display, including orientation (e.g., track-up vs. north-up) and viewpoint (e.g., 3D perspective vs. 2D planar), as discussed previously. For 3D displays, he also compared the egocentric (or immersed) view vs. exocentric view (in which the aircraft itself is included in the field of view). Aretz’s review concludes that a rotating, track-up map display results in greater flight guidance accuracy (e.g., Aretz, 1991; Olmos et al., 1997) as well as improved hazard awareness and global situation awareness (e.g., Olmos et al., 1997), compared with a north-up display. The egocentric view results in better local guidance (e.g., Wickens and Prevett, 1995) and may be especially advantageous for very low altitude flights (Hickox and Wickens, 1996), but it results in poorer global Situational Awareness (SA) than the exocentric view (e.g., Wickens, 1999). However, the exocentric view results in less precise judgments along the line of sight and greater ambiguity about object placement, relative to the egocentric view.

In summary, the choice of map display (including frame of reference, geographic scale, and information displayed) depends on the tasks to be accomplished. Based on conclusions from the research sited, this author selected a track-up 2D moving-map display and egocentric 3D display (consisting of conformal, scene-linked, symbology superimposed on the FFOV) for the experiments presented in this thesis.

1.3 Overview of thesis organization

The thesis organization follows the evolution of the project itself. The section following this Introduction is dedicated to several focused interviews that were held with F/A-18 pilots experienced in the use of cockpit moving-map displays. These pilots led the author to

investigate methods by which advanced cockpit displays might support the transition between internally dominant and externally dominant tasking during air-to-ground missions, such as CAS and FAC. The interview discussions can be seen as an informal task analysis of these missions, from the perspective of utilizing a map display or HUD.

The third section of the thesis proposes a cognitive process model developed from these interviews and discussions. The model is intended to help visualize how a pilot senses,

available to support successful decision-making and ultimate control of the aircraft. The proposed model can be used to identify possible deficiencies in the pilot’s decision-making – particularly during the integration and correlation of multiple sources of information – and to suggest appropriate remedies.

The fourth section presents several hypotheses for improving pilots’ correlation of information in support of two primary phases of an air-to-ground mission: flight guidance and target

acquisition. Three experiments are proposed to test those hypotheses. Detailed experimental methods and statistical analyses of results are included in this section.

The fifth section summarizes experimental results and presents relevant conclusions. A list of references and four appendices are provided. Appendix A provides the source code for a computer program, developed for this thesis, that links Microsoft Flight Simulator (running on one PC) with a moving-map program running on another PC. Appendix B provides the

calculations for the root mean squared error (RMSE) between the actual and planned flight paths and the RMSE between the actual and planned altitudes (measures of flight guidance

performance and workload). Appendix C provides plots of experimental results for each subject. Appendix D includes a copy of the consent form completed by each subject prior to participating in the experiments and the debriefing statement provided to each subject following their

2. P

I

Three F/A-18 pilots with experience using cockpit moving-map displays were interviewed in October 2001 at the Naval Air Warfare Center in Patuxent River, MD. The purpose of the interviews was to gain insights into how these pilots used the displays to assist them with various missions and to solicit recommendations for improvements to existing displays, based on the pilots’ mission requirements. The primary interview topics included 1)!general information about the pilot; 2)!tactical air missions that are (or could be) supported by a moving-map; 3)!how the pilots acquire and maintain SA; and 4)!recommendations for new moving-map display functions. Follow-up telephone interviews with two of the pilots were conducted in March 2002 to focus on specific issues raised during the initial interviews, including the degree of flexibility a pilot has in executing air-to-ground missions and a proposed transition phase between

“internally dominant” and “externally dominant” guidance in many air-to-ground missions. Table 2-1 presents questions that were used to focus the interview sessions.

2.1 General Information

All three pilots were experienced flying the F/A-18 fighter jet, with 1200, 800, and 700 hours as pilot in command, respectively (table 2-2). Most of their flight time was in the older F/A-18 C/D “Hornet” (C designates a single-seat cockpit, and D designates a dual-seater with both a pilot and navigator). The F/A-18 C/D is equipped with a moving-map system that displays scanned aeronautical charts with pilot-selected overlays of certain mission and intelligence information, such as approximate locations of threats and targets.

Two of the pilots also had limited experience in the newer F/A-18 E/F “Super Hornet” (E has a single seat; F is a dual-seater). The F/A-18 E/F is equipped with newer avionics and cockpit displays, including the Tactical Aircraft Moving Map Capabilities (TAMMAC) system. TAMMAC can display satellite imagery, in addition to scanned charts, along with more advanced mission graphics such as height-above-terrain (HAT) and clear line-of-sight (CLOS) overlays. HAT uses the Digital Terrain Elevation Database (DTED) to generate shaded overlays depicting terrain that is higher than the aircraft’s current altitude; CLOS uses DTED to highlight

regions that are obscured from the pilot’s view due to intervening terrain. A Terrain Avoidance Warning System (TAWS) will also be incorporated into a future version of TAMMAC.

Table 2-1. List of questions used to conduct focused interviews with F/A-18 pilots.

General Information

• Name, rank, contact information

• How many hours do you have as Pilot in Command (broken out by aircraft platform)? • With which (if any) cockpit moving-map system(s) do you typically / currently fly?

Tactical Missions Flown

• With which tactical missions do you have experience?

• If you could think of a particular mission scenario, from the point of view of using the map, what would be one of the most challenging scenarios, and how would you use the map? • What features or functions were missing from the moving-map that you could have used in

this situation?

Situation Awareness (SA)

• Specifically what does SA mean to you? • How do you acquire SA (step by step)?

• How do you use the current moving-map display to support your development of SA? • Is there anything that could be done to the moving map to further enhance your SA?

• What features or functions are missing from the current moving-map that you wish you had, for improved SA?

Advanced Moving-Map Display Recommendations

• The TAMMAC map display offers several new functions (e.g., HAT and CLOS overlays). Are all these new functions useful to you?

• What other functions would you like to see incorporated into a moving-map display? • Does your current moving map provide these functions?

• If so, how could they be improved? • If not, how would you implement them?

Pilot Flexibility

• How does a pilots’ level of flexibility (e.g., in ingress / egress routes) impact their ability to accomplish their mission?

• What, if anything, limits your flexibility (in terms of both internal and external guidance)? • Why is your flexibility limited (internal / external guidance)?

• How could this flexibility be improved (internal / external guidance)?

• If a target is not located where you expected (e.g., if you had bad intelligence information), what do you do next?

Transition Phase (Internal fl‡ External Guidance)

• There seems to be a transition phase in many air-to-ground missions from "internally dominant guidance" (i.e., relying primarily on cockpit instrumentation, including the map display, for flight guidance) to "externally dominant guidance" (i.e., relying primarily on the outside world view for target acquisition) as you approach a target or region of interest. Does this make sense?

• If so, can you expand on this idea?

display, other instruments)?

Over the next several years, the TAMMAC program office plans to incorporate additional advanced functions that will rely on the implementation of a vector-based map database. Some of these new functions include the ability to selectively add or remove features from the moving-map, query the map for more detailed information, and allow text to remain upright as a track-up map spins (Williams, 1998; Ruffner and Trenchard, 1998).

Table 2-2. Military flight experience of interviewed pilots.

All three pilots had additional experience flying other aircraft without a moving-map display (500, 590, and 3800 hours, respectively). At the time of these interviews, the first pilot was the F/A-18 project officer for the TAMMAC program, so he was very familiar with both existing and planned moving-map capabilities for this aircraft. The second pilot had extensive combat flight experience, having completed 24 combat missions (some as a flight lead) during Operation Desert Storm in the early 1990’s. The third pilot (who is also a medical doctor) had flown during both Vietnam and Desert Storm, and had over 4500 total flight hours in the military. He had considerable experience testing night vision goggles (NVG) and moving-map displays during training and test flights.

2.2 Air-to-Ground Missions

The three pilots identified several missions for which they would use (or had used) a moving-map display, including Forward Air Control (FAC), Close Air Support (CAS), Combat Search

and Rescue (CSAR), Deep Strike, and air-to-ground targeting in which the target is stationary (e.g., a building) or moving (e.g., a tank).

The second pilot discussed some of the differences among three F/A-18 air-to-ground missions:

“With CAS, you circle overhead as the FAC talks you through the situation. [In this mission] our friendly forces and enemy forces are both right there. We would have established airspace superiority, so we own the airspace.

“CSAR is a lot like CAS, but there can be more threats. We have a route, like in deep air strike. We need to keep an eye on the rescue helicopter [which flies at] 100 kts max. The “rescort” [F/A-18 rescue escort] job is to provide air cover for the helicopter and [must be able to] visualize the route. The guy to be rescued is sending secure GPS coordinates [to the CSAR].

“Deep Strike or Deep Air Support occurs in the first days of war. We’re hitting the enemy, so there are no friendly forces. We’re taking out a factory, etc., and may not have established any superiority. That [means there is] time pressure. I need to know if that’s an enemy tank without a doubt. On Deep Strike, … there is a lot of pressure and stress. If I look around and the target’s not there, I will high tail it out of there.”

Despite the danger of Deep Strike, pilots identified the FAC / CAS mission as one of the most difficult air-to-ground missions, because of the close proximity of friendly and enemy troops, the need for close communication between FAC and CAS, and the CAS’ need to correlate numerous sources of information. The pilots also stated that the FAC / CAS mission could specifically benefit from a well-designed cockpit map display. The first pilot described it this way:

“One of the hardest things to do effectively is CAS for friendly troops, but that [mission] is not going away in modern warfare. We can have all kinds of smart weapons, and shoot things off ships with GPS guidance, but the guys on the ground are still the ones who will be very dynamically involved and needing very flexible support. That’s where moving-maps [are important], with the ability to cue the pilots in terms of, ‘What is the terrain? Where is the FAC? What do things look like? Where is the target?’ That’s when it gets really important. We’re talking the new war on terrorism – very mobile, very tiny, very difficult. That kind of stuff will be next to impossible unless you have near-instantaneous data.”

Figure 1-1. Illustration of a sample FAC / CAS mission.

According to the second pilot interviewed, a given FAC / CAS mission might be supported by one or more FAC-Ground (FAC-G) troops, FAC-Airborne (FAC-A) pilots, or a combination of both, who are in radio contact with another pilot providing CAS (figure 1-1). Perhaps a FAC-G sees an enemy tank rapidly encroaching on his or her position and radios instructions to guide the CAS pilot into the area and fire on the tank. To successfully accomplish the mission, the CAS pilot must correlate information from several sources, including the FAC-G’s instructions (which are brief and cryptic), the view outside the cockpit, Forward Looking Infrared Radar (FLIR) imagery (if available), and the moving-map. Table 2-3 summarizes the tasks that make up a typical FAC / CAS mission, information needed to support this mission, and several problems with current displays, based on a combination of discussions with the interviewed pilots.

2.3 Situation Awareness (SA)

Endsley (1995) has formulated a widely accepted, three-tiered definition of SA as "the perception of the elements in the environment within a volume of time and space, the

comprehension of their meaning, and the projection of their status into the near future." A pilot must obtain different elements of awareness (e.g., hazard awareness, navigational awareness, automation mode awareness) in order to achieve SA under different circumstances (Endsley, et al., 1998). Therefore, understanding how a FAC / CAS pilot acquires SA should be an important step toward identifying the appropriate tools to help that pilot perform a FAC / CAS mission more effectively.

All three pilots interviewed for this thesis emphasized the importance of acquiring and

maintaining SA during flight, but particularly during the dynamic and mentally taxing military air-to-ground missions. The second pilot gave his definition of SA and explained how he acquires and maintains it:

“Overall, SA is knowing where you are, what’s going on around you, and projecting into the

future what needs to happen, what you need to do with the airplane, and what potential problems

are developing. SA is where it’s at; especially for an F/A-18 pilot and especially for a flight lead. “Let’s start with why SA breaks down. For me, SA breaks down when something distracts

me, something that wasn’t expected, … something weird. For example, if communications are

breaking down, because someone isn’t on the frequency they’re supposed to be on. With a distraction, you end up fixating on that problem, trying to solve that problem, at the expense of all the other things that are going on that you need to be aware of, such as: Where’s my wingman?

Is he flying in proper formation? Where is the tanker? Who’s on the radar display? Are we in the right piece of sky? What’s coming up next?”

“Basically, [I regain SA] by solving that problem or ignoring that problem and figuring out if that thing that’s distracting you is really that big of a deal. If it is a big deal, you have to focus on it. You can’t keep scanning for other things; you have to fix it. It’s tough. SA is what separates an F/A-18 pilot from anything else. Especially if you’re an F/A-18 flight lead, you have to have ‘God-like SA’. That’s what we always say! You have to know what’s going on all around you.”

Numerous aspects of SA are required for the FAC / CAS mission, as the third pilot described:

“There’s navigational SA, tactical SA, communication SA, all of it together is knowing what your mission is, where you are, what your job is, where the bad guys are, what’s happening behind you, what your weather is doing, the whole thing. The [important] thing to me about SA is not to get behind the aircraft but to stay ahead of the aircraft. If you get behind at any point, your SA is also behind.”

The third pilot emphasized the importance of communication and trust in establishing and maintaining shared SA in missions that involve more than one aircraft:

“Your SA is somewhat divided between yourself and your lead [aircraft], and your tasking [is divided] between your lead and your wingman. A lot of a wingman’s SA is derived from the lead. [The wingman must] assume the lead knows where he is … If your lead flies into a mountain, you should be a heartbeat right behind him!

“Talking [with your lead] is essential: paying attention to where you are, knowing when to get your weapon system set up, knowing what kind of maneuver you’re going to conduct. When you’re not talking to each other, SA becomes more difficult. Even getting four people to agree on what the mission is [is difficult]!“

Managing inevitable change in a mission scenario is crucial to maintaining SA (third pilot):

“I don’t think I’ve ever flown a flight that went exactly as it was briefed, because change is a part of the [mission]. I’ve been on a lot of mishap boards; for most of [them], change has been an integral part of the mishap. This [includes] changes in the flight crew or schedule or launch time or target, from the time you sit down and start planning until the time you’re out on the mission, any change in there can effect things. You have to understand what a change means and then how to [manage] that change … that’s all a part of SA.”

The cockpit tools that can help a military pilot acquire SA are mission dependent, as the first pilot explained:

“It depends on what the mission is, and what I’m doing. If I’m doing air-to-ground, then I’m worried about the ground. If I’m flying air-to-air, I’m just worried about where I am geographically. Sometimes, [for air-to-air] I just need a little box on a blank screen. But the rest

of the time, I will have done some preflight planning, so I’m going to have a [preconceived] picture of where everything is. Depending on how accurate a study I was able to do, I will want to visually correlate where I am into my big geographic picture. The first thing I’ll do is look around and, if I can, see where I am. I’ll correlate that with my mentally derived picture of what things ‘should’ look like, based on my preflight study. Then, I’ll ‘come inside’ to look at the map and determine what I could see outside that I can correlate to my map. Then, I can start using the map to judge distances to features. How I perceive distance will vary, based on my altitude and depression angle. If I’m really high, I’m going to look like I’m close to everything. If I’m flying at night, NVG can really destroy distance perception, because I can see stuff so far away.”

2.4 Advanced Moving-Map Display Recommendations

Current military moving-map systems can display scanned aeronautical charts, satellite imagery, reconnaissance photographs, and potentially anything that can be displayed as a digital, raster image. Future moving-map systems will also be able to display features from vector-based geospatial databases, such that individual map features can be manipulated, allowing the display to be customized (Lohrenz, et al., 1997). Given the vast amount of information available, cockpit map display designers must take care not to present too much, resulting in an overly cluttered display. In addition, designers must weigh the benefits of cartographic flexibility (i.e., the ability of pilots to customize their own displays) against pilot workload. A good cockpit moving-map system conveys critical information concerning navigation, threats, and targets in a manner that is easily interpretable under often-stressful conditions. The pilots who were

interviewed for this thesis shed some light on which geospatial information is most critical for air-to-ground targeting missions such as FAC / CAS. This section summarizes their responses. 2.4.1 Geospatial features of interest

Typically, the pilots use a moving-map to help them find geographic and stationary threats (e.g., rising terrain, power lines, towers, etc.) and navigational “handles” to use as lead-in features to a target area, if the map is of a large enough scale. The second pilot gave some examples:

“In the F/A-18, the primary information I need from the map is a ‘God’s eye view’ [i.e., 2D plan view] of what’s going on: boundaries of areas I’m trying to stay inside of, routes – low level, strike route, ‘yellow brick road’ – and the sequence of points I’m trying to fly. Other information currently displayed on the digital map, like terrain and airfields, really is not required for the F/A-18. The digital map is good for correlating critical items (routes and waypoints) with what the pilot is seeing outside. For example, while following the yellow-brick road, if you know the next waypoint is supposed to be over an airfield and, in fact, you can see the airfield symbol on the map under the waypoint symbol, it gives you a ‘warm and fuzzy’ [feeling that you’re on the right track].”

The third pilot explained how he has tried to correlate the information on the map with his view outside the window:

“For example, if you see a power line going across the map, you know darned well you’re not going to see the power lines in real life. But, you might be able to see the pylons on top of a mountain, if you have enough contrast [e.g., dark pylons against a light sky], so you would look for the pylons.”

When asked what features or functions he would like on the moving-map display to better support FAC / CAS missions specifically, the second pilot discussed the need for more descriptive, mission-specific symbology:

“In a FAC type mission, we have CP (control points) where the CAS airplanes are hanging out, IP (initial points) to fly through to get you on the right attack axis, and targets. We can have multiple CPs and IPs. When we’re making charts and putting a plan together, we have different symbols for those different points, but you can’t do that on the HSI [Horizontal Situation Indicator display], which is not good. They’re all the same “waypoint” symbol, which is a little circle. [Drawing on our charts,] we use a circle for a CP, a square for the IP, and a triangle for the target. It would be great if you could code those on the moving-map.”

“The FAC-A needs the Forward Edge of Battle Area (FEBA) and the Forward Line of Troops (FLOT). Those different boundaries would be good to display in different colors. Right now, … it’s all green and it’s just a line. It would be great if … the squadron could standardize it … and decide to always put the FEBA in black, always put the FLOT in green, CPs will be circles, IPs will be squares, targets will be triangles, and it would be standardized. Then you could just look at your map and it would make a lot more sense. This, for me personally, would be huge. Use colors and different symbols; go through the critical things that we use, which are waypoints, airfields, and maybe terrain, but no roads or names of all the cities. There’s too much stuff on the map, and [the screen] is too small.”

The second pilot discussed ways in which several aircraft flying in formation might use the moving-map display. When asked how important it would be for them all to be viewing the same information on the map display, he answered:

“It can be [important]. Or it can be that one guy is responsible for one thing, another guy responsible for another thing. It depends on the situation and on the flight lead; the top gun will decide. There is a division of airplanes doing different things, looking at different pictures. I might say, ‘Wingman, while we’re on this cross-country flight, I want you looking at emergency airfields.’ Then if one of us has a problem, he’s ready and immediately can say, ‘Hey, the nearest airfield is 170 deg at 30 miles’. But there are also missions where everyone needs to be looking at the same [map]. It must be standardized somehow, but still be flexible. For example, it would be nice if the flight lead could tell the map to show him exactly what his wingman is looking at.”

2.4.2. Correlating ground and air perspectives

The first pilot talked about the difficulties of correlating the FAC-G view from the ground with the CAS view from the air:

“This is a big problem – there are very different perspectives from the ground vs. the aircraft. A 50 ft rise on the ground is a hill; in the air, it’s nothing! A FAC could data-link his position [to the CAS, who could] then slew over and elevate [the display] to see where the threat is that [the FAC is] talking about.”

This problem with conflicting perspectives must be addressed before reconnaissance

photographs (which can be displayed on the moving-map display) are useful in a FAC-G / CAS mission. The first pilot described having to mentally rotate an image taken from one perspective on the ground to match his own perspective from the air:

“To provide me with a picture in the cockpit, [reconnaissance troops] went out and took a picture of the target from the ground. It was a very clear photo, so I knew what I was looking for. Here’s the problem: I am not coming in from the same perspective! [The picture was] taken from the ground, while I’m going to see it from overhead. Now I have to correlate my ingress direction … with whatever direction this photo was taken. They gave me a picture of the truck from the side, but I’m flying in from the nose of the truck. It looks absolutely nothing like the picture, so the picture actually confuses me at this point, because now I have to correlate in my mind what that truck would look like if I were coming from it straight on.”

However, the same pilot predicted that reconnaissance photographs taken by a FAC-A pilot might be very useful to a CAS pilot, assuming the FAC-A knew from which direction the CAS would be flying. In this scenario, the photographs would serve as a visual link to facilitate communications between the two partners:

“When you get into FAC[-A] and CAS, then [a picture] is important. What’s important to me is where my people on the ground are – and can I correlate that with my mental image of the whole battlefield. Pictures taken by the FAC[-A] and linked to me would be excellent, in that respect. … Pictures taken from his perspective I can kind of correlate [to my view]: he says he’s here; that’s the view; that will put him about here; OK, that looks right to me.”

2.4.3. Timeliness, accuracy and trust

The first pilot discussed the importance of real-time imagery or photographs in a dynamic targeting scenario, especially when enemy troops are located very close to friendly troops, as in the FAC / CAS mission. If the picture is too old, it becomes useless:

“As a FAC-A, it’s hard to find the friendlies and correlate them to the bad-guys when they’re only _ mile apart, since I can cover _ mile in about 2 sec! You’re just zipping by going, wow – who’s who? If I’m going to bomb a bunker, [that’s not a problem because] the bunker’s not going to move. But if I’m talking about vehicles, scud-launchers, how do you find and kill mobile targets? That’s very difficult to do because it’s very dynamic, and if the information is

more than 20-30 minutes old, it’s probably wrong.”

Trust – or a lack thereof – in the pilot’s various sources of information was a recurring issue discussed in these interviews. If the accuracy and timeliness of the information being provided (e.g., by a moving-map display) cannot be trusted, then the display is of limited use. If the pilot trusts the information, and the information is, in fact, correct, then the display can be of considerable use in making critical decisions. However, if the pilot trusts the information, but the information is actually wrong, then the display can contribute to potentially fatal errors. For example, pilots have been taught not to rely solely on maps (paper or digital) for terrain avoidance, as the first pilot discussed:

“There have been several cases of Hornets flying into mountains because they relied on the INS [Inertial Navigation System] map data when, in fact, there are a lot of ways to get that wrong. Aircrews are very hesitant to use the map for any kind of ‘no-kidding’ terrain avoidance. But, as we graduate into GPS-aided positioning, then I think that will start coming about. … If you align the INS at the wrong spot, it’s going to send you to the wrong spot. You can align [the GPS] wrong too, but once you get up it will correct itself. Unfortunately, many folks are coming from a background without a GPS-aided jet to rely on; they haven’t really started briefing or training folks to use GPS for any kind of terrain avoidance. I don’t know when that will happen.”

Another issue related to trust was mission urgency. If a mission is critical and urgent, the pilot might be willing to take more risks by trusting the information provided, illustrated in the following account by the first pilot:

“A FAC[-G] positioned in a valley … wanted to [shoot] a couple of tanks because they were rolling in on his position. He was calling for air support, but it was an overcast, really bad day. The weather was bad enough that … normal patrols weren’t up. So, the Marines launched with a section of F/A-18s. … There weren’t any air controllers in the area. The [F/A-18 pilots] had the waypoints, so they looked [at the map], saw where the FAC was in the valley, and just oriented themselves off [the map], let down through the clouds and were able to prosecute an attack on the tanks. So there was a situation where these guys are – despite the fact that they probably shouldn’t be – relying on the map for all their SA and to orient them into the target area. That’s probably a worst-case scenario: close-air support, where you don’t know where people are, or the kind of help they are going to need. You’re launching, but you haven’t been able to do an effective target area study, so you don’t really know where you’re going. That’s about as bad as it gets, where now everything that you’re going to accomplish you’re going to do based on information that’s available in the cockpit. It obviously has to be concise, readily interpretable, and accurate, so you can accomplish the mission.”

2.5 Pilot Flexibility

This author theorizes that the level of flexibility a pilot has (or believes he or she has) in

executing a mission will greatly influence how much (and what) information should be displayed to the pilot. A mission in which the pilot has less flexibility (e.g., if the pilot cannot deviate from a predetermined flight path) would seem to require less assistance than a mission in which the pilot has more flexibility in determining ingress, targeting, and egress maneuvers. A pilot’s flexibility on any given mission is closely tied to his or her perceived threat level, as the first pilot interviewed for this thesis explained:

“[My missions are] fairly scripted, depending on the threat level. With a higher threat, I am less flexible, less likely to deviate. A lower threat level means more flexibility. A strike mission is very scripted; FAC/CAS is more flexible. SCAR [Strike Coordination and Armed Reconnais-sance] is the most flexible: the aircraft has an assigned area of responsibility, or kill box (10x10 or 20x20 miles) to look for targets. The pilot will call others for support if he finds a target.”

When a pilot arrives at a target area, the target is not always where it was expected to be. There are numerous possible reasons for the problem:

• The target may have moved;

• The target may not be visible to the pilot;

• The pilot may have made an error during ingress to the target area;

• The pilot’s ingress was correct, but the pilot may be searching in the wrong place; or • The pilot’s intelligence information may have been wrong.

The pilot’s perceived threat level, mission flexibility, confidence in his or her sources of

information, and confidence in his or her own performance all dictate how the pilot would handle this situation. The first pilot related his own experience:

“My immediate response is usually, ‘I’m wrong. I need to fix this’. Or, the target moved. I need to go back and compare my mental notes with what I saw [i.e., mentally tracing the route again, from waypoint to waypoint]: the field was right, the barn was right, the map looks right. I need to validate my environment before talking to my FAC again.

“If the threat is low, I’ll stay overhead. With a high threat, I might leave and go elsewhere. If I expect to get shot at, I’ll go away. I might abandon that search and start over again with a new set of references. Although if I am a bomb dropper, I’ll rarely start over.

As military capabilities evolve and threats change on the global scene, pilots must be flexible and adapt old strategies to meet new challenges. The first pilot recalled:

“Folks who flew in the [first] Gulf War [say] it started one way and ended up a completely different way. Tactics change; you meet the environment and a lot of things that previously had been held as paradigms turned out to be wrong. Like going in low. We lost quite a few aircraft [in the beginning] because of outdated tactics. Pilots were trained … to go in low, thinking that would work, since it always had in the past. Aircraft were getting shot down right and left going in low. So everyone turned around and went in high, and that worked fine. It just depends on being able to adapt to the threat environment. But you learn fast! You can’t train to any one particular threat scenario, because you don’t know what you’re going to face next.”

2.6 Supporting the Transition Between Internal and External Guidance

A recurring theme throughout these interviews was a need for the pilot to alternate between internal cues (the map and other cockpit instruments) and external cues (the world outside). This process has been described by other researchers as navigational checking (Wickens, et al., 1994; Schreiber, et al., 1996), in which the pilot compares characteristics of the map with

characteristics in the FFOV to determine whether or not the two are congruent (Aretz, 1991). Navigational checking may not require any other action if the pilot determines that the flight is still progressing along the correct path, but the process nevertheless continues throughout the entire flight (Conejo and Wickens, 1997).

One of the topics of this thesis is investigating how this transition between internal and external guidance might change as the pilot nears a target area, and how a given display design might support this transition between the flight guidance (mostly internal) and target acquisition (mostly external) phases of a military air-to-ground mission. This topic was discussed with two of the F/A-18 pilots.

The second pilot described looking for “funneling features” on the ground to cue him into the target area. These could include one or more linear “hand-rail” features, such as a road or river, or a series of “attack points” that lead to the area of interest. The pilot described studying reconnaissance photographs of the target area (if available) and trying to correlate that information with what he saw outside or on his FLIR. Correlating the outside view with a picture can be very difficult, as described earlier, depending on the orientation and perspective of the target in the picture, the time of day the picture was taken, etc.:

“You study the pictures carefully and look at the maps, but the area can look completely different from what you thought. Vegetation, especially, can be a different color. Reconnaissance photos are usually in black and white, which provides better detail; FLIR

is monochromatic [green]. The good [CAS] guys can orient themselves to the map or picture, with [varying] daytime and shadows. [It would help] if you could process a reconnaissance photo to show how the area would look at this time of day and this time of year, [since] the photo could be a year old or a day old.”

The first pilot talked about the hazard of relying too heavily on preconceived ideas about what the target or target area should look like:

“I might have some prejudice from past experience: I’ve been here, I know this area, I have a mental image of this experience. What [the FACs] are describing may match what’s in your head and you don’t think you need to validate it. This is dangerous. [On the other hand,] if you have no idea what they’re talking about, you’ll keep digging until you’re comfortable. If you have prejudice, you expect something based on what you’ve heard – like ‘high terrain’ – but he may be saying a hill, and you’re expecting a mountain.”

This pilot went on to say that good training methods should help alleviate this problem, by teaching FAC and CAS troops how to communicate more effectively with each other. The primary responsibility for good communication in this case lies with the FAC, who must describe his or her surroundings (and specifically, describe the target within the context of its surroundings) in a way in which the CAS can readily interpret.

The second pilot described how a FAC-A pilot and CAS pilot might use a high-contrast ground feature (visible to them both) as a common waypoint and unit of measure. For example, they might agree on the length of a nearby runway as 1 unit of measure. Then, the FAC-A could instruct the CAS on where to fly:

“While the CAS is circling, the FAC-A says, ‘There is a river running N-S. OK, see that bridge crossing the river? Now follow that road east for 3 units’ [i.e., 3 runway lengths]. It’s hard to refer to some [ambiguous] features, like a big field. The FAC-A has a map of the target area, and he’s talking to [FAC-G] guys on the ground; he flies into the target area and looks at the battlefield for a while. The guys on the ground give him a detailed brief, discussing the situation. When the [CAS] aircraft shows up, [things get] very time-critical. They don’t bother referring to their maps [any more] – it’s too time time-critical.

In this case, the FAC-A was already in the target area, but he still alternated between inside cues (the map) and outside views (the ground) until he understood how best to instruct the CAS pilot when the time came. The final transition occurred when the CAS pilot came into view, and both pilots’ focused on the scene outside. The second pilot suggested that highlighting important

lead-in features and other cues (but not necessarily the target itself) would help improve these transitions between inside and outside:

“JSF [the Joint Strike Fighter program] is looking at using a ‘virtual world’ to combine database information and maps in a heads-up or helmet-mounted display to funnel the [CAS] guy in. On TAMPS [Tactical Aircraft Mission Planning System], he can type in a ridgeline to view, and the ridgeline would be highlighted. OK, here’s the river, etc., to cue the pilot. You can’t do this with targets, because they might be moving.”

Without the benefit of highlighting, a target can be very difficult to find. Because of this, CAS pilots historically have relied on FAC-G troops to manually mark the location (if possible) with flares (Gunston and Peacock, 1989). Obviously, this method is highly undesirable if there is any chance for enemy troops to extinguish the flare – and possibly the FAC! Virtual highlighting of a feature (e.g., on a map display) eliminates that opportunity for an enemy to detect the FAC. Three types of virtual highlighting have been shown to improve target acquisition performance by reducing search time: luminance, flashing, and color (Conejo and Wickens, 1997), with color generally accepted as the most effective method (e.g., Yeh and Wickens, 1997).

Unfortunately, highlighting can hinder the pilot if the highlighted object is not actually the correct target – i.e., if the highlighting is invalid (Conejo and Wickens, 1997; Fisher, et al., 1989). Highlighting is most valuable (in reducing search time) when trying to find a target against a complex and unfamiliar background, since it reduces the need for the pilot to scan the entire complex visual field. Conversely, highlighting is least valuable when finding a target against a simple or familiar background, since the pilot’s scan time (without highlighting) would be much shorter, and the potential benefits of highlighting would be less significant.

3. C

P

M

The focused pilot interviews helped to identify several channels of information that a military pilot uses to make critical decisions during air-to-ground missions, including the pilot’s external “head up” view of the real world; the internal “head down” view of cockpit instrumentation, and audio input such as radio communication (figure 3-1).

Figure 3-1. Primary channels of information used by pilots during air-to-ground missions.

Figure 3-2 presents a more detailed cognitive process model developed to help visualize how a pilot might sense, perceive, integrate and correlate the many – and sometimes disparate – sources of information to support decision-making and ultimate control of the aircraft. As shown in this model, the pilot receives visual information about the real world from the external view through the cockpit window, which is perceived in three dimensions. Different features in this view exist in various focal planes (near, midrange, far) and can be hidden or partially obscured by weather or terrain. A HUD, superimposing navigational and guidance features on the external scene, may supplement this real-world view.

Figure 3-2. Cognitive process model of how a pilot senses, perceives, integrates and correlates various sources of information to support decision-making and ultimate control of the aircraft.

The pilot receives additional visual information from “internal” cockpit instruments, such as gauges and displays. The military pilots interviewed for this project utilized a moving-map display, which presented geospatial abstractions of real-world features in a 2D plan view. The pilot must correlate these 2D abstracted map features on the internal display with 3D real features in the external FFOV to successfully guide the aircraft to the intended target location and, ultimately, acquire the target.

In addition to multiple channels of visual input, the pilot receives audio information from radio communications with the ground (e.g., FAC-G), other aircraft (FAC-A), or a crewmember (e.g., the navigator in a 2-seat aircraft). In the case of a FAC-G / CAS mission, in which the CAS pilot is attempting to acquire a target based on information relayed by a ground-based FAC, the pilot must integrate and correlate his or her own visual information with the FAC’s verbal description of the target area. The success of the CAS pilot’s mission is highly dependent on 1)!how well the FAC describes the target and surrounding area and translates this viewpoint (an immersed view from the ground) to the pilot’s overhead perspective; and 2)!how well the CAS correlates his or her own views (internal and external) with the FAC’s verbal description.

As shown in figure 3-2, one more channel of information with an impact on the pilot’s mission is the pilot’s own expectations – his or her expected mental model – of the target area and the target itself. This mental model is developed during mission planning, which would have taken place prior to takeoff or – under tight time constraints – while enroute to the mission area. The pilot’s mental model is continuously updated as new information (internal and external) becomes available throughout the mission. As one pilot discussed during his interview, a poor mental model (based on erroneous, preconceived ideas about what to expect) can mislead the pilot and thwart the mission.

The proposed cognitive process model reinforces the importance of a CAS pilot’s skill in

transitioning between internal and external guidance to reach the intended target area and acquire the target. A central hypothesis for this thesis is that well designed, shared cues between a CAS pilot and his or her FAC will improve this transition process and contribute significantly to the success of the air-to-ground targeting mission.

4. E

4.1 Introduction

Three experiments were designed to better understand which visual cues in a moving-map display or HUD can best help a pilot 1) navigate to a given target area (the “flight guidance” phase of a mission) and 2) search for, find and identify a target (the “target acquisition” phase). The author hypothesized that in the case where the target is stationary and in a known location, the most important cues would be those that support the pilot’s guidance to the target area. If the target had moved, the pilot would still need flight guidance cues, but he or she also would need target acquisition cues, which should be shared between the pilot (e.g., CAS) and the ground controller (e.g., FAC-G) to facilitate communication between the crewmembers about the new target location. Sharing information between civilian pilots and ground controllers has been shown to improve situation awareness for both agents, foster collaboration, improve negotiations and decision-making, and reduce operational errors (e.g., Farley, et al., 1998; Hansman and Davison, 2000). Presumably, similar benefits could be gained by sharing relevant information between FAC-G and CAS in the military air-to-ground scenario.

The first experiment, Experiment A, was a quick test of whether pilots would perform better (during flight guidance and target acquisition phases) with a turn in the route or on a straight-line approach to the target area. This experiment primarily served to establish baseline performance measurements and workload assessments for each subject during the two mission phases. In both cases, subjects were presented with a moving-map display that included the following graphic overlays as visual cues: planned flight path, target location, own aircraft location, and a breadcrumb trail (updated once per second) to show the actual flight path. A simulated HUD was also provided, with graphics embedded in the pilot’s forward field of view (FFOV) to represent the planned flight path and the target location. (Detailed descriptions of the moving-map and HUD are provided in the Methods section.) Flight guidance performance was expected to be better on the straight route, especially for subjects with little or no previous flight

experience. However, target acquisition performance was expected to be better with a turn in the route, since the turn itself (i.e., the flight path displayed on the moving-map and HUD) would provide distance and heading cues that should assist the subject in locating the target.

Experiment B concentrated on the moving-map display and provided no HUD symbology. This experiment tested how well pilots would perform with three different displays:

1. Combination display: displaying both a topographic map and graphic overlays of the flight path and the last known target location,

2. Map-only display: displaying the topographic map with no graphic overlays, and 3. Overlays-only display: displaying the overlays with no underlying map.

Detailed descriptions of the displays are provided in the Methods section, below. Pilots were expected to perform best with the combination display, since this option provided the most cues about where the flight path and target would be. Better performance was expected with the overlays-only display than the map-only display, theorizing that overlays provided information that was more mission-specific, while also presenting a less cluttered display, than the map. Experiment C focused on the relative merits of the HUD vs. the heads-down moving-map display to support flight guidance and target acquisition. Pilot performance and workload were compared for the following cases:

1. The no-HUD case provided subjects with a moving-map display (with both the topographic map and overlays presented in Experiment B) but with no HUD symbology. This case required subjects to perform a forward mental rotation of the track-up image into the FFOV in order to align the two views, as described in Aretz (1991).

2. The HUD case included the same moving-map display as in the previous case plus the HUD symbology presented in Experiment A. In this case, no forward rotation was required since the required symbology was already present in both the map and the FFOV.

3. The HUD with north arrow case included the same moving-map and HUD symbology as the previous case, plus a north direction arrow over the target location on the HUD.

Each subject flew two missions for each of the three test cases: one mission in which the target had not moved from its last known location (marked on the moving-map and HUD, if provided) and another in which the target had moved a short distance. At approximately 2 km from the target, the experimenter (acting as a FAC-G) told the subject (acting as CAS) whether the target had moved and, if so, how far and in what direction. For stationary targets, the biggest

improvement in performance was expected between the no-HUD and the HUD cases, with very little (if any) performance differences between the HUD and HUD with north arrow. For targets that had moved, significant performance differences were expected between the no-HUD and

HUD cases and between the HUD and HUD with north arrow cases, since the addition of a north

arrow should help the subject orient him or herself within the scene and find the target. The following sections first describe the methodology common to all three experiments, including a description of the participant population; apparatus; design of the moving-map display, HUD scenery, and targets; experimental procedure; and dependent variables measured. Each experiment is then described in detail, including an overview of procedures specific to the experiment, hypotheses, and results.

4.2 Methods

4.2.1 Participants

Twelve volunteers were recruited from the Naval Research Laboratory (NRL) detachment at the Stennis Space Center, MS. Nine subjects were male, and three were female. Two subjects had a pilots’ license, although neither had flown in several years. Three other subjects had experience with flight simulators. Subjects ranged in age from 24 to 52, with an average age of 37.8. 4.2.2 Apparatus

A Dell Latitude C800 laptop computer running Microsoft Flight Simulator 2002 (FS2002) generated the flight scenery for all experiments. The display resolution was set to 1024 x 768 pixels and 32 bits for true color. A Dell Inspiron 4100 laptop computer running the Navy Portable Flight Planning System (NPFPS) FalconView program generated a moving-map at the same resolution and color settings as the FS2002 display. The two laptops were linked via serial cable with software (in Appendix A) developed to perform the following functions:

• Extract positional information from FS2002 in real-time, using a Microsoft Netpipes socket; • Convert the data to National Marine Electronics Association (NMEA) 0183 compliant GPS

strings; and